Available online at www.sciencedirect.

com

Journal of Vocati nal Behavi r 7 1 (2 0 0 7 ) o o 34 03 58

www.elsevier.com/locate /j vb

Emotionala n d p e rs o n a lity -re la te d e c ts a sp q of persistencareerdecision -m akind i cu lties t g a b, Noa Saka , ItamarGati * a b D ep artm enof Education, t The H ebrewUniversi of Je rusalemJe ru s a le m ty , 9 1 9 0 5 Israel , D e pa rtm e nof Psycholog, The H ebrewUniversity Je ru sa le mJe ru sa le m t y of , 9 1 9 0 5 Israel , R e ce iv e d2 2 May 2007 Availa ble online 22 August 2007 Abstract

This d i focused on exam inin g the persiste ntaspe cts of ca areer d decisionmaking studyculties,usin g the Em [Saka, N., d Personality-related ree r (in ecisio nC m aking Diculties sca le (EPCD ; otion al an Gati, I.,decision- m aking di culties. p re ss). Emotional Career A s s eality -re late aspe cts of career four Kelly, K.R. m easures and person s s m e n The contribution of & personality d Journal of t]). ste e m tra it a n xie ty , a n d id en tity sta tu s to th e general indecisiveness, re erde cision -m aking cultie sw as te ste don 7 4 7 stu d en ts, , p re d ictio nof p ersisten c a se lf-e t u sing s at th e b eginndesign. e academ icschdi year had less con den cein the ir sc o re a longitudinalclose to making indicated that individuals winto high EPCD ingof th Results a decision about the m ajor ith w hich they ool choice and w ere le ss at the e nd of th e year.The m od erate w antedscore adm ittedfour personality m easuressup ported th e validity of th e to b e and th correlations betw een e th EPCD Im plications e counseling EPCD. for and future re se a rch re d iscu ssed . a 2007 Elsevier Inc. All rights reserved. Keyw o rds: Careerdecisionm aking; otio decision-m ality-related Career culti Career and person di culties; aking indecision; Careerindecisiveess; Em n nal careerdi es s c a le ;EPCD Career indec ision is often m aniested as di cu Marmarosh, & f ed w hile m 1 9 9 3 c a re eKrausz, decisions (Chartrand, Rose, Elliot, lties encounter Caldwell, aking; Gati, r- related q Th is re se archw nce o Butler,R. Kelly, Tali as supporte dby the Israe l ScieTal,Foundati n. We thank Ruth Chani Etengo,Reum aLilach Sagiv, i Goldblum , Moshe Tatar for their h Kevin ents an earlierversionofGadaarticle. Shiri Kleiman, this ssi, Naom and Valentina Izrailevitc , com m on * Correspoding author.Fax: + 9722 5 8 8 2 05. n 4 E -m a il a dd ress: itam ar.g ati@ huji.ac.il (I. Gati). 0001-8791/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jvb.2007.08.003

N. Saka,I. Gati / Journalof V o catio n a B eh a vio r (2 0 0 7 )3 4 0 3 5 8 l 71 341 & O sip o w , rs9 9 6 ;L e o n g & Chervink be 9 9 6 ;O sip o w , C arn e or during the 1976). no, y, & usually refe 1 to proble cance of to 1 addressedpriorvocati nalBarak, decisioIt ms that m akingpro- cess.Th etheoretic need careerindec and career counselos (e.g., Betz, signi ians, resea ision in to o psychology has b ee n highlighte d by rchers, r 1992;Campbell & Cellini, 1 9 8 9 ;Gati e t & l., insle y ,;Je984 ; n &n D ille y ,0 1 ; 7 4 ;Le o n g 8 ; Chervinko, 1 9 9 6 ;O si o w , p 1991; Rounds a T 1 9 9 6 1 p se S a to s,2 0 1 9 la n e y ,1 9 8 & S Tinsley, 1 9 9 2). In additon, the distincindecisiven esscon tinu es developa sign i cat ision and i tion betw eentem po rary, to be mental indec focus of m ore ch ro and ervasve onal-de nic, p edu-i cati vocational De-Boeck, 2 0 0 2 ; cision research (Callahan &5 ;Kelly n h as, 1 9 9 2 ; re e n u Gantoe ijs2 0 0 The term develop- m entChartrand, & Jo w d y , 1 9 9 b Genu sedLee, 2 0 0 2 ; e r s, and m Cohen, l indecisi has genera & to referto S e normative vocatio develop a phase that is reso lly e easily for m ost 1). on th naln g , 1 9 9 3 ;Slaney, ment lved fairly young adults (Betz & S e rl i 1 9 8 8cisiveness invol es m ore pervasve, severtoand chronic di cult ision, ca reer Meldahl st mental ies in in d e - ; deci- & Muchinsky,1 9 9 & Ini contra e, develop sipow , indec99). making v(Meldahl 7). Muchinsky, 1997; O career e n e sssions arenly presentin a sm allergroup of individ C are er indec levelsis app isiv t uals,19 characterized by high iu sed of n seof personal identity (Cohen etthe self and the choiceaproces3 ; ing about al., 1995; Chartrande t l., 1 9 9s, and a d & Muchinsky, 1 9negative think se anxiety, 9 Meldahl 7). Career decis ion-m am g the lties sig stem from em i colleand persona ng lityrelate d sources are aki on di cu orethatni can tdi cu otonal ge stude m lties nts face (Amir & Gati, 2 0 0 6 eAmir,1 9 9 6 ;S a k aKleiman, press; Santos, 2 00 Amir, subm cu s are conside d p re ss & 1). These diitted for public re ation; Gati ; t a l., Gati, & to tba l., in insevere;Gati in form aion- elated di cu ltie by care ercounselors e e more th an t r lties (Gati, Amir, & Tal, 2 0 ci- sion-m are p erce theas havin g im p o s. In additi n, ces e di clientspreven tr 7), ived counseling proces conseq uen som the cultes caree rtant for de 0 and aking and conclusive decisi and require long-terminterventi ns that ices (Gati et a l., 2 0 0 e ons o o t i scope of careercounselors an d caree s to enlin g centers servfrequenly exceed th r-coun se 7). The personath e plated n t re se a rcdi cu goal of lity-rere s- e deci- sion h a lties hance ourating their relationwith three w unders tanding of emotional and by persona m e asu s and focusng on their pers investig lity re i istent,chronic a sp ecs . t N rous ies ave ined th e relation sbetween career indec lity a and um e n s stu d th eho the exam ., Kelly &one hand, 5and th ese vario us p ersonand ; vioral charac r (e .g ision indecbehaes1 ;,Slaney, teristics, on the Lee, 2 0 0 ;Leong & Chervinko, 1 9 9 6 San toisive0 0 on s, 2 1 9 8 Th e va ria s stu d 8). ble ied in c lu d e se lf-e em a n d self-condence (Kishor, 1 9 8 1 ; ste Santos, 2 0 0 se l -e ca cy(Taylor & Betz, 1983), lo cus of control (Taylor, 1 9 8 anxiety 1), f 2), (Fuqua, Seaw o rth, & N ew m an,1987), personal and vocatio al ide ntty (C ohe n e t al., 19 9 5 ; n i Santos, 2001), 1. Em otion aland pe rso nality factors associat w ith care er de ci ion-m ak ed s ing dicultie s The taxono t developedby Saka et a l. (in my i person- , ality-relas that w ere consi on-making p re ss)fo cu seso n th e em otonal and career decisi re sea rc variable ed w ere located ntly found diculties. Relying on com m on h ste and analyz d b e correlated with previous to for their source, and teristi s, and the similarity in the ty pe of interventon needed. Theindecision indecisiven ess e charac taxon y om was develo c p throughth e distinc betw een eoretical i sideations and em pi ical th con r r testing ,stic ped rop Anxiet ,interplay -Concept three m ajor. clusters of diculties: andView s, osed ay and tion Pessim iconsists of diculties related to dysfuncional percep Pessim istic Views Self am ongand Identit The y cluster biasesabout the self and the world. The Anxiety clusterinclud esd negatve t tions and icu lt iies cogni ive th e anxiety provoked by the decision-m t in vo lving, w hich m ay preventor inhibit the decisionaking process s. its potental and i outcom s e making proces 2 . T h e e le v e n diculty c ateg s orie istic vie s. The rst m ss istic v of the m el, proce i s. of Pessimcateg oiew The rst ngajor cluster ie w w hichodthePessimstic View s,consists pe self-e im cacy, s aboutis ssrefe to a tion rs low degthreofor s h r decisi re e e aki indiv ee percep that he career isp ro on-mTh e of co ca te goy p essimthe view idualsc tivw orld of r e decision-making incapable percep ing regard thorough and b.g u fee ecca rere ce s. ve se carry ndtions out a occu r istic s(e o .,tth w a re es a work a lly interesting). Fin ally, the third categ oy pessim refers to overly negati ing pations a re re u ls cont ol refers to th e in div uals sen se of an exte r istic rnal locus of in divid aove r the proce s, the nal choice, and/or th e o utcom s. view s abou t th e r id control e Anxiety. The secondm ajor cluster u t the m odel, labeled to feelings y consi ts of cat- egories. The rstanxiety a b o of th e proce the decisnxiet aki of sstre ssfour ssrefers Aion-m , ng proces ,an d anxietyevoked b y exc es-siveactually begin ab o u t th e p roc es. This catego or anxietyinarising just prior to t in e rfe ren ismning im is ticv ie w sa b o u tth e p ro c e s T h e p n a tu from p e ss ctio s ry is sim ilrr more fo c u s e don the cogniive percep la ttea is wcontent but dieren sest the feel- in gs ofnes lpla b ility to en g a g ss. that tion of o he in process,during the actual p roces . The second cate- go ry hile the forme r focu on essnexiety du e ein the ss and stress th e e mce ga in ty in v o lv e d in cho osng inc des three d im en an unce rt to (a) e s lu u n cerrainty about th e future , (b) an xie about being in an undecided ainty : i sion s of un et rt ty

N. Saka,I. Gati / Journalof V o c a tio n a B eh a vio r 1 (2 0 0 7 )3 4 0 3 5 8 l 7

3 43

w hich m ay bem anifestedin diculties in e x p resing con sol idated beliefs, lu es, refe ces,and va p ren life goals. It also refersto an uncrystalized and unstab l le Emotional and Personality Related Aspects of Career Indecision Pessimisti c Views

Anxiety

Self and Identity

abou t the proc ess

abo ut the wor ld of work

about the one's control

abou t the proc ess

about uncerta inty

abo ut the choi ce

about the outco mes

gene ral anxi ety

self este em

uncrystal lized identit y

conflictua l attachmen t and separation

Fig. 1. Th e taxon om yof em o tionl an d person a ality-rela teda spects c a re e rdecision-m of aking diculties .

3. Th ego al of th e p re s e t re s e a n rch

The m ain goal of S a ka et a l.ntstudyss)in to ex amne the w e exam i th theoretical the w as tw i m odelproposed onsratinpre sepersistenasp ecs w ays.First,validity ofnedethe m odels validitySby nd ,m b yin vestigted (in e correlati to of the dicultie s as(ree re sen itng the de w e t ga the th p re t on s betw eenthe EPCD m pasu re d the by EPCD. eco o del)an d m easu res gene ral inde c proposed m These m e a- sureofe re chose nas the ore isive n anxie ty,that ide ntitype ct to support the constr ct validity of th eess, serelem odel.In addition status. w u s tically lf-estee m , vantcorre late sand w erege ne ralm e asu of inde c isiv e - n e ss ,in this stu dy wprop osed on threem e asu s ex ed to a w e re hypothere d to be assoc e focused that oncep t and ize tit . Fu tu reiateddes sh oe categos o n in v e sig a tin g h o ws th rese s ries inclu t See a sue s a re a s soIden dwith th e otherwtwo th u ld fo c u lf-Cr stu i ith m ajor clusters.It dedin the clu ter eof m idate all threemc ia te y w not stu val m e a sue s u sedajorclusters tebelow al mth eorized relachns assinggfeasible to for th e an drth e valid ity m exp lic a u sed. , w e easu with sever the easu res e atio ,in a onle the dy.For am EPCD scales res S elfn- cept in vocatirch oice is ofte n v ie w1 9 5 an expression of a centr l role in -estee A caree onal terms (Super, ed as S e lf e s te e mplays an individuals m. self-co ing ones self-concept (Chartrand, Robbins, - Morrill, & B o g 3). a 0), actualiztend to choose care ersth at will allow them to actua gs, perce a s p eop le and en hancetheir feel- ing s of self-w orth. Further their 1 9 9 potentialnegativecorrelati b e tw e e n lf-e ste e m inde c more,lizeany studieshave m lower ones ived found a the higher onesindec on ision (Kishor, se and ision: the se lfe ste e m, 1 9 8 1lations be tw 0 0 n ; Wul & S teitz, 1 9 9 total EPCD scoreexpectedone hand, to s , 1 9). We therefore i corre;eS a nte em, 2 eethe other. clus also hypothes the three We te rs and the ized that the ms, onclus negatve the of Se lf and s ptand Id en lf-e s on w ould show the highestcorre ajor m . ter C once tity lation w ith s e lf-ete e s Trait anxiebeen foundtrait anxie i anly th com m enhereascareer an d career al to ty d correlated, t to a h ave consi - tentlyty. G enerfound be ne gatv eivelye corre itmw(Hartm anxie choice B lu m , s ty indecision ghave been inko, 1 9 9beMeyer & W in er, 1993;W anb an,Fuqua, & to 6 lated 1 9 8 5 ;L More ov er,San tos(2001)founposit anxietyand general indec & Mess to be erg uchinsky, 1992). e o n & Cherv- lated, and ; trait anxiety and vocational identity to b e d trait highly positively corre We thereforeexpect d posi ive correlat isiven een the neg atively scoresand the th ree clus score on th e t correlated. e ions trait anxie betw total EPCD r. Again, w e hypothesizedthat the m ajor clu s ofand -Conceptand ter s, one hand, Self ty, on the w ould show th e hig hestcorrelaton with trait anxiety. ter Iden tity othe i Iden tity statu Gener self and ein, Deven s. a identityrelated variables,such as personal be - is, voc a - tional identity varialbles (Blustn e ap proa h & Kidney, 1989),relationun d toand corre status and career choice.diculties is the parading the w ere fo s betw een late ith career indecision O c to stu dyi su gg t iden tity d w classied indi- viduals into one of fo u r id e n itty sta tu se s ed th e Marcia gm (1 9 8 th ey m ade m ajor d ecision sin their lives and the ir de greeof es n m itm a sis ) who o by ent of h ow 0,decisions: (a) achie ve d identity, (b) identity forecl com o b , to those osure, (c) m orat riumtoand (d) diused ith careerindec m ora identity. Bothision (e.g., Blustein dif- fu sed1id en tity w ere foun d torium and et be corre lated B s, W e there w expect d positive correlati ns be tweal. th e total in & Savicka 1994). fore e o e n 1 9 9 ; risb EPCD totaland diused cluster score on th e on e h and, also hypothes res of s, and th e m ized m ora ajor cluste r of Self-Conceptstatus, on the othe r. We the highest easu that torium and three the m identity statu s midentity . and Identity would show correlaton i with the e asu s re G eneral indecisi a s. G ener l situations (Frost & aS h o tes, 1a s 3 ; Gati e t al., venes ety of indecisivenessis m n ifes d 9 9 a diculty in a m aking decisi ns in d escrib s inde c o 1 9 9 6 ; C rite s (1969 all vari e of lifeisive pe rsonshe thertheyw who s eatm to little as individuare als e hav e diculty in (pp.)). sorts de signcance making3053 isiven i ion betw een indec Further cisions,w lity characteri of gre eor oan 06). rch has associat ca reerin d e csion . To te s t our h ypoth es that the EPCD estab-suresml ore ess more, prior resea and persona stics.e alished n g Th pro aspec and persisten td icult i is stab ts of le ies, w carr for specic m rs, thus d m to ch oosea ied Israel, before subm itti to univere cation ou t a. follow-up longituinal anal- ysis. In apply m ajor individualsprogra their appli ngms w hich sities a secondch an ajo Israeli havingw ho wish to form In m cefor indivi uals ost universities there are pre-aademuniver ity but hav e not taken their high- chool matriculation oer applys, or the icto incre toc w s ase their ch anceof s d exam titive ant rs like com p uter scien being adm itted(in particular for courses com pe g rame s can substitue for their ce or chool exam) grades that did not y by taking highly whose for th e program of t d ajo high- psycholog this p rogram invol es s qualify g tim e (a year)and m oney,as w ell as Participating in their choice. sig- ni v in ves p rogram , th e stude are expe cte d to reachcan tacadem ic eort. By the e nd tin of the ajor(s)to apply for. Therefore, studen whoa well-dened prefe nts whatthey are w illing to comm it to b y the end of this ye ar, despite the gre atabout m ts do not have a decision m ajor rred eort that inves en t required du rin g this year, m ay be in a m o re pers and isiv etm ss. istent state of indec ne 4 . Hypoth eses (i) T he correation betw correlati ns scoresof the EPCD and a m easuesof generalinde-cisiveness egoeen o relativ statuswill be mthe total. betw eenthe EPCD and m easu of self-e elyl rre steem, anxie and be ty, will identity high, and oderae t (ii) less progress in highdecision-m resat process, (b ) will beefurther aw ay sho w Individuals with the EPCD sco the beginning of th y e a r:(a ) will from m aking a choice , and (c)EPCD aking s .a ch oice, will b e less con den t abou t it than individuals with low if th ey m ad e sc o re 5 . Metho ds 5 .1 .Participa nts Seven red and en studenIsrael the pre-a in cadem study.Three t programshund rty two re e forty-sev (4s7 .2 % )in ts m a le sa icipated in .8 ic w preparaory th hund not of fo e - th largeste irs e an a g ew ere 22.4partn d 31 .7 (5 2 this)hundfe male s; red stu duniver ities w a n % re red and 2 0 did and reportof th esex ;th e n ts w e re born sin Israel; =5 9 4 00%) hade gradua their stu - d e t m (SD 6 8 (9). Five fty-nine (76%) and received a m atricu ted from high school, or had nishedbut did not docerticate,w hilerece e a diploma . lation w ell enough to 7 2 iv 0 % )h a d not (1 nish high school Longitudinal analysis pl ar,w the 747 part sam e e. icipants ho lled nses of EPCD 3at out th e beginn lled out the EPCDOf e w ere able to m at hwthea n sw e redthe follow-up c th en respo th e 9 5 (5 %) stud en wing of the school yw as attribu o n dto technical problem s (e.g.,3missing fo a table time questonnaire.The attrition i ts hoto m atch th e rst andr th sec nd-timand uestionn aire), missing data, informationendancein class in the secondround ofe data collection,due to various e seco q and non-at rese t factors (e.g., rand , arm y serv ice, illness,or person s reason To veri y s). f attrition w as the rve5 partcom pare d thelled out the alon th e naire at bothof data om w e icipan who EPCD sco req ueston rst round that the collection for 39 the partcipan w ho w ere absentduring the second round of ts ts i rounds of data collection and diere nces w ere found betw e e n th e s c o re s of th e s e two i data collection. th e total EPCD score nor in any of its th re em a jo rc lu s rs (t(7 4 5 ) No groups,neith 5 ) in 0 .1 2 ,t(7 4 5 )= 0 .4 6 ,a n d t(7 4 5 )= 0.6 0 ,re sp e c - e ly ).Furth er er te = ere.0 9ere 4 die re nc in the de m ogr th 0 w ,t(7 = no es aphic varia s be tw ee n tiv ble the groups (a more, 2g e : = yea rs of e d u c a on : t(7 1 1 ) ti = 0 .7 2 ; s e x : v (1 , N = 7 2 7t(7 3 3 ) 6 1 .1 6 ; )= 0 .6 ). 5 .2 .In stru n ts me

Th e em tionl an d perso a i The consi m of th e EPCD ois descrbed in nality careerdiculties scale(EPC D).EPCDdevelop ento5 3 details ). ite m age, and yea rs of ed ucati iin Saka includesgeneralbackgrostatem sts of n: the questonnaire et al. (in p ress Th e sex, s.The rst page of 11 diculty cate goie in g pages includ e 53t, th einform ati ts on. The follow s. For each statem e und participan en each ts, repreasked to raof th e d e g re eto w hich th r statem en t escri e te the n w ere sen ting on not de scribem e to 9 d e s c rib e m e w eld a higher rating indic them on a 9-point scale (1doe sdiculty). Two valid- ity ite me w e reinclude din bed sca leto ensurethat s l; higherlevelof s th e con side ates a individuals replied only w hen somethingreadinga p - p eite m s me, and I dont m ind ring respon r ms(I expect tions a re re a lize do r n ogood h pract n s to and of th ese vathe ir am satis d after properly t). The the ical util- ity w hethe se yrep o rtedin eS aka et al. (in p re ss; th e rationaleand te s ts of this kind of a lid ity item s w ere ) re ss validity items is describedin Amir e t al. (in p ) . Psych etric properties w e p found m b e w ere q u5 Cronbach-alpha om ad a consis- tency reliabilitiesin therere se n t to p ively.e The,te: ,.8 8 ,a n d of 5th e 1internalthree m ajor clus- ters and Pessi istic View respect th e W .8rld.9 5 W o rk ) to .9 2for th esca es the total scal , ssa le e t l ran gedfrom .61 (ford A n x- iety a bout th e C h o ic ). A p revio lities y.9 (for 1 al., in m abou e o reliabi stud (Saka et o f us Anxiety about th e Proce ss a n const p re s support d clust rs and scale s. Spof ci ca , conrm the proposed analy s) e the ity e- th lly and structure of thadequacy ofruct validotheszed e EPCD ical atory fact r internal o supporte d catego s e the e rie grouped into the hyp th eorete sis e diculty .9 8 , a n d GFI = .9 7 ;Saka etth reiin majorthe inte rs model= of 5 7onlethen e p res s; cluste rcor-re lationsam ,NFIv= (RMSEA .0 .9 8scalescore sin th e pre s e n tsam p are p resen in AppendixA) . ,CFI = al., g 11 le ted Frost inde cisi ne individu ve e (FIS; Frost extent s,1993). th ey FIS is ith 5 -ite mselfS how Th e reportingatheirw hichss scal . be inde c the& Eachto w hich ratedagreewa 1 statem en ts regardsc le in tendencyto als rate isive . item is on a ve-po t scale in ran gin g from 1 str ngly disagree to 5 str ngly agre The scoringis base don two sub sca le s o o e. labeled Fears about Decision-Ma king and Positive Decisio n-Makin The intern g. al consi ten cy s (Cronbach-al pha) of th e FIS total sc o re w a s .9 0 in a studentsam p (Frost & le Shows, 1 9 9 it w a s .8 9 for the Fears subscale an d .83 for the Positivesub scae in a 3); l com m un ity sam p (Steke le tee, Frost, W in cze,G reen ,& D ougl ss, 2000). In the p re se nt a sam p (N le = 7 4 7 )th e Cronb ach alpha inter nal-con sisten c y lia blity e stim aes w ere .8 0 ,.6 8 ,and re i t .8 5 for the Fear Posi ive, and total FIS, s, t resp ecively. t R o s e n b e rg lf-e se stee msc a le(RSES; Rosenbeg, 1965). The RSES, one of th e m os r tused scal s in th e a s se e ssm e n tof se lf-e ste e m a n to s , 0 0 co n sis ts 10 items exp resing (S 2 1), of s ve positive and ve ne gatv e v ie w s of the se lf. Indiv i iduals are ask e d to rate th e deg re e to w h ich th ey agreew ith each item on a 4-pointLikert scale (1stron glydisa gree to 4 stron agre Th e ne gatve ly orie nte d gly e). i items are the nre ve sed, so that a highe rscore r indicate sa highe rle v e lof se lf-estem. The total scoreis com puted the m eanof all ten e as items . Satisfacory psychom tric propertiesof th e RSES hav ebe e nre porte d:its Cronbacht e alpha internal-consistency reliability w a s a b o ve.8 0 ,a n d its testreest re lia blity w a s .8 2 t i (Flem ing & C our ney, 1984 t Trait anxie scale (TAS, from the State trait anxiety inventory ty (STAI) Spielbe e r, rg Gorsu ch, & Lushe 19 7 0 We used the H ebrew versionof th e TAS, which is part ne, ). of th e STAI. This scale a s s e e s relatively stableindividual dieren c in anxietyss es proneness. Th e scalecon sistsof 2 0 sta tem en a n d th e individual is a skedto rate th e d e g ree ts, to which h e or sh e g en e ra experien s th esee motion s, o n a 4-pointLikert sc a le(1a lm o st lly ce

never to 4al ost all th e time). Nine anxie m ty-abs t item sare rever ed, an d th e total en s scoreis c o m p d a s th e m ea no f a ll ite m s , s o th a th ig h e rs core re p re n ts a h ig h e rd e g re e f ute a se o trait anxiety. E xten d e d o b je civ e m e a s u re e go id en t of tity statu s (EOMEIS-2, Bennion & Adams , 1986). This scaleevaluates indivi duals identity status basedon the m odel propose d by Marcia (1 9 8 0 Its go a l is to estim ae th e d eg reeof identity consoldation for four ). t i statu s e s (achieved identity,identity forec losure, m ora torium , and diused id entity).We u sed th e Hebrew version of th e m oratoium an d d iu sed id en r tity p arts, so that th e sc a le consisted of 3 2 items (1 6 for each identty status). In div uals w e re a ske d to ra te th e d e g re eto i id which the y agre e dw ith e ach state m e nt a 6-point Likert scale (1 str on ongly disa e to gre 6 stro n g ly g re a e)..

D ecision stat (DS). This m easu w hich is a variationof the O ccupa us re, tional Alternatives Q uestion(OAQ, S la n e y ,1 9 8 0 ; Zener & S chnue lle ,1972), asks individ uals about their careerplans direc tly, to deter ine their cu rren tstag e of the care erd ecisio -m ak ing m n proc e ss , and c a n b e u s e d to asse the ir progres (Amir & Gati, 2 0 0 Gati, Saka, & ss s 6; Krausz, 2 0 0 1 Gati, Kleiman,Saka, & Zakai, 2003). ; 5 .3 .Procedu re About 4ms eksafter the beginning study on en w ere ts th e progra w ewcounselo ued b y the of th ecschool year,studion-m aki ,in fo ed th e rs about ng rm th e and by questits (in groups d istrib o u t 2 0 3a )d u rin g herscareer dcecis e uate stu den towas onnaires ere three grad ts stud en l; 16 stud en of a b t dto part5 resear o n e and th e thess s.Part booklet blank of left la resea icipation optiona clas ts refuse icipate and either rch or left the sroom (about2% ). TheEPCD andts w FIS, given a bookletRSES, the TAS or th e EO M EIS . Finally, they questi onnaires, w -2 s th e participanthe ere phic either the with the included the DS.hich consi ted of and lled out a e EPCD a rst, and the order of the o th er scal s w as All participans onnaire that t lled out th demogr the questi e R SEScoun terbalaced among 6participans. Of. Ident participans, 246 lled ou t the )w as 262 th el;n 7 0 0 747 on by nam e (on the last page t , optiona and 2participan t 3 th e EOMEIS-2 icati es,w hich allow edus to m atchthe book TAS, (9 4 questonnairesin the rst and secondrounds of data collection. %) ts reportedtheir nam lets with the i The- second adm in hich included the EPCD and th e DS, w as 4 w ee ks later.th e istration of the questi nnaire took p la ce 2 distribu o follo w ts (in grou of a b o 2 0 3 5 ) u rin g o n e of th e cla sse s, s in the rst round of let, ted to The studen up book ps w w e re able to dm at ut a dataen . collection. We ch the rst and second book lets for 3 9 5 stud ts 5.4. analyse s P re im inary l

For each p RSES, TAS com p u tedth e total sco e EOMEIS-2 (whichever th e EPCD and the scores ofarticip an tw ean d the for o scale sof thresof m ajor clus and th e FIS,lled out). We com-com pute d corre s tw am on g th e EPCDe (its total te rs and they ajor the lations the EPCDs 11 N e x t w e also thuted the FIS, RSES, TAS and thre EOMEIS-2 score,s. No scal s. p e scales), score three m e clu st rs,as found for th e FIS, TAS, or the tw o E O Mtw -2 s c ale s(all Fs <e Therefore, e and o scal1). order eect wlts are reported acrossthe various EIS the resu administration orders. The decisicom binedthe sixre originally included sixthe individ on stat m easulevelsinto four to reect levels (Gati e t a l.,ein0 th e us 2 0 . 3) Howeve r, we sta g career decision- m aking p rocessaccord g to theSpecica (Gati & Asher,which: in Choice. PIC m odel the uals level, 2 0 0 1) Prescre ning, In-dep explor tion, and en in g of altern ativesstag rst represe nte d th a corres e te mto being bd o nret e v e n h ae a general dire lly, and e, wI o n ly h a v ea ponded nts (1 ) I efo o the Prescr e by th e sta e v respon ctio n tranition be tw e e n (2) as g en eraedirection. and second level, Ex p Prescr etsin(3stage The the In-D e pamcor- a sm ading to erof spec eln th ong loration ll n um b the represented by the stag w as s e, th e sta te(4 ) n g mI am deliberain g ic occupa tions and m before I conside ri decision The third level, corre like ic to plo Cothe r I a ) m a ke mng a t spec tion, but w ould to occupa options as repre se nte d y sta te . e nt know w hat I a m in te re ste d , ex t wreu ldlike sponding b u the hoice stag w m y choic by the th em rI level indic e, in o to conrm e. Finally, fou th ishedth e process,represented by th e statePIC nsI g ated reabosformhat idwsix h ad nm to th at th e in divI ual m ajor in.ption srationale for categ othe is outlined ss uGati andw the ouldlike to The into th e four using ries m e t ta a e in tranu t Tal (in p reo rigin al resp onseo ss; this ) categ oization wof found useful in empirically process (Amir & Gati, individuals in r die ren t . es as th eir careerdecisi n-m aki dierentiating am ong 2 0 0 6 ;Gati e t o ng a l., 2 0 0 stag 3) 6 . Results 6 .1 . C o n str uct validity an d d ive e n t rg

T ab andp resens th e ad devia r tions, and th e correlat ions o f Iden EPCD le 1 atorium scal m ea n s,sta n dith the FIS, RSES, leftmost columiusedthe total its t tity and Mor- e m e three m ajor clusters w es of the EOMEIS-2. The tw o TAS and the D ns of Table 1 pre se n tth ans 350 N. Saka,I. Gati / Journalof V ocationa B e hav ior (2007)34035 8 l 71 Table 1 M eans,standarddeviation, and the correlations s betw e EPCD and the validity en questionna sa ire Scales M SD N EPCD Pessimis Anxiet Self Tot views identit y.6 0 Frost Ind ecisivness Scale e 2 .6 5 0 .6 3 7 3 9 .3 6 .4 7 .5 6 R o se n brg se lf-e ste em le 3 .4 0 0 .4 9 2 4 6 e sca b .2 0 .3 0 .6 1 .4 4 Trait-an iety scale x 1 .9 4 0 .5 2 2 6 2 .2 6 .3 6 .7 0 .5 1 Ego identityscaleDi sed 2 .6 4 0 .6 9 2 3 6 u .1 9 .2 3 .2 8 .2 8

Ego identity scale 2 .4 8 0 .7 2 2 3 6 .3 6 .4 2 Moratorium M 4 .7 8 4 .7 3 SD 1 .3 8 2 .0 2 a All co rrela tion s re sta tistica lly n i cnt (p < .0 1 ). a sig a b The correla tions with EPCD a re n eg a tive s e xp ecte d . a

.4 8 3 .5 6 1 .4 4

.5 0 4 .3 2 1 .3 4

and standa ep tfor self-estee e semic a sue s. As caeszedse en ,all correlati ns a re rd ns o lated positive,exc deviatio of th w h e as hypoth ni b e , is negatvely corre m, h, r i with th e EPCD. The correlat of the total EPCD score w ith v alidity m e asu s range from ions the re .28 for th e covarance, inable, th e initial EPCDstatus towgroup w as the of th e yea r vari ble, i the decisi ar s the d depe nde ntvari w hich on status on th e covarate. This end independed w a s th e a the and the initial decisi deci- sion as diculty i th e three group srew as nt that ally w analysis vea statistic le diere nce in the 3 8 )= 8 .3 4 ,p < .0 0statu s am ong initialdierenc am ongthem (F(1 , nal signicant (F(2 ,.0 0 1 2 ). 1 ), beyo the nd es 2 3 8 )= 9 .4 3 ,p < In su m ry , th e se n dng s p rov id e fu rth e rsu p p fo r o u r c la im th a t th e EPCD ma i ort indeed taps into the m ore stable ,e ndu ringcom pone of caree rd ecisionnts makin g d icultes. i Spe ci ca lly, the ndi gs indic n ate that a high EPCD sc o re p re d ics le ss p ro g res in th e t s decisionm aking processand low er choice condence, after a signicant period of time in which natural crys tallization of career plans should take place. 7 . Disc u s sio n The goalslof th e p resen stu d y w ere tig diculties and to t emotional andvalidity of the tax on t y proposeto inv e s ate the characeristics ofer nal and lated o in th e personaity-re careecareerdecisi n-making et al. (in th e EPCD furthoti exam e em personalit themdU sing ome asues for dby dicult isiveness,self-escale develo y-re late . rdecision-m generalindec an d pre ss)for steem o general aking Saka ies, ped to m ty, identity diusionm e asu re r ora ,analysis, anxie and torium, and conductingnom y tand th e w e obtain to resu lts supp ortin m w erevaliditybetw the the total EPCD score and the ed taxo a longi udinal Modecluster high s, on the ong h and, foundtheof e a- sure s of generalin dec EPCD. ationse th e and thre e ra te scorecorrel ty, an d self-e m eene other. Further ess, ego-identity, g en eralanxie steem, onakin g a careerdecisi than those more, isiven ts with a low EPCD sc o re. ad vancedless to ward s m th on stu den w ith a high EPCD score G eneral in d ec isiven es. m e asue of inde c s on eenthe Frost ecisiven e S cale ss (FIS), whichcaree ge neral The correlati e c isb etwne ss,and s .5e Ind2 after correction a focuseson is );r-relateddlation rshow sd th atisive em otio a th 6 total EPCD, w hich w for atte ation th is nu a re correlated, iculties inized, wio ne the generaland(.6of indec corre as th m akin g , nal traitpersona isiveness. lity-re lated diculties corre hypothes ith The lowest (r lation between the FIS an d th e EPCD for this nding e Pessimstic i View s oiessin this = .3 6 ). One the statem en repre- senti as foundin thth EPCDat th e clu ter possible explanation w g th em for e is th w ere categ r specic and e m band n highly en w ere gclusterThe nding that thecareer context, w he th e FIS w as w ith e dded w ith in th ts highestcorre e reas all th e FIS sta tem-C oncep t an de ra tityclus of th e EPCD (r = .60)lation of th at th is clus ts en l. th e S elf ts th e m oreIden ter m ay sh ow ter repres n e g eneral asp e i y-related career diculties,thehereas the and hence its of em otonal repre personalit e lower-sp eci i cts correlati ter and sen ts more asp ec of w diculties, Pessimstic Views clus with the FIS w a s th carees t. c ts on S e lf-et st rs,.As w ell a s s e m a hypothesize all d, ation s lat b etw en isre ea sue an d th e th th r m ajor,clu e en egatve. This show s th atco rre b io n seenthisem easumity-relatedaspec ree correl s co re r decision-smaking allicultiestheare etw al and personalf-e the total EPCD em oti on l and ts ca ree w e re as i could b ed xpe ct the h ighe st correlationse th e e m Scale wof higher when of ste RSE is low er. Furthth e clu sterof Self-C ce p ta nd Id en titysca leof th e EPCD. ed, as w ith erm ore, on e Trait anxiet . As hypothes y ized, all correlati ons betw een this m e asu and th e re total EPCD score and th e th ree c lu stersco re sw ere positive and statistic ally signicant. The highestcorrelat ion w as again found with the clus ter of Self- oncept and Identity, C and m ay be e x plai d by the fact that this clus includ esa sub-scaleof generalanxiety. ne ter Still, a correlaton of .5 1 betw e e nth is m e a sue and th e to ta l EPCD scorem ay in dicate i r th at an x- iety is o ne of the com p o ents of pers n isten t career d ecisio -m akin g n diculties (Germe ijs,Verschu eren , S oenen s, 2 0 0 . & 6) Ego ntityScale and the correclustersof the EPCD w sed siive subscale of th e ty scal . e iu Ego Ideidentiow ever,The threelationssbetw een the dthan p oidenofand statistically t titythe m or signicant. H e Ego Iden tity statu s.Thtently lowcorre e re those the diu sed atorium they w ereconsi e hig hest er lation of subsc ale of th e Ego Identty Scale w as w ith the clusterof Se lf C once pt and Id entity th id en y; sub scale ofbe attribu this m ay ed identit .ted i to the fact that this cluster includes a su b-s cale tit of uncry stalliz y The m ul iple regrebe stpanalyor revea otthat general indec t w as the ssion re dict of em ional and perso isiveness, aspecs red ses led as m eats of u by th edif- culties, as m easuredby th e EPCD. In additi n, all th reed t caree r FIS, o nality-relae personality

variablesself e ste e m , tra it a n xgeneralindecisivenes to the prediction of th e ie ty a n d id entity sta tu s (moratorium) contributed (5 % , 3 % , and beyond cre l EPCD score signicanty, e nding th at theseaccounteds, varirediced th e resu e ase in -for v ariable s, resp e tadm inistation% in th e EPCD vari- ables m onthsnce, for the thre cively).Th 1 6 of also pa later) support the t lts of th e secondte n t charac of em ot r ter chrovalidity of th e EPCD a s a m e asu of (alm ostnality- relateddiculties and is ional and perso6 the nic, pers re such diculties in c a ree r decisionmaking. T tity of th e jo h The Id e n e th ree m aayr diculty rg e individfairly EPCD. rs free persona decision-making e m cluste rsof the context- cluster of Self-Concepta ns da a lity charac teristi d c cluste r that m EPCD lations w ithstheuals rnal m e asure s ere re erhighe s. Therefore, predis e dicultie s,hypothesizedthat there m ay be ato ppo iste nt ca the the clusters, and as its corre pose e xte tem e ral order w ithin w t it m diculties invol ing a non- ohesive self ay be that in parallel (orv c l i exist (Gati e t al.,even lead to) dicult and ay b e lack ative of essty status m ay ies m an uncrystalized ident for making to prolong careerdecisi ns but alsoinvolving in dicof indecisive. sposition 1996). Th ese dicultiesto be ge n-e rally readin predidecisiona not only o 7 .1 .Im plic ations A ssessng em otio i nal and nality-rela im aspec aking dicul- ties has both theore perso perso ted plica ts ofBetterunderstanding of tions. ca ree r decision-m th e relationsbetw ethe basis for a com prehesivettheory of ca re pers een em otonal and practicaln i tical and nality-relaed factors and er indecisiven s , istent career diculties m ay b es delin eating es of the career various m aking nents andTh e proping explicat on various sta relations am ong decision-com po proce s. Specically, a their eectm ay m se rve as theg rst ste p forisiven constructing such a theory. through ose d tax ono y d ies m ore thorough understanding decidindec of ed and ess could be achiev ed who dier in beha longitudinal stu com p ar g cs such as postpon their choice for long period changing m ajo indec vioral characin es,and returning to personal couindividuals, al tim es. s, teristiun ing isive nsel sever rs m anytim ing Cong thlingrs t plicati in assisting ssm en.tTheclients loca ou nse e im ste p s ons. The asse them of re foe, career-dec source s of the ision diculties is am r isiven ting the dicultie ca ree rcounseling.Accurate and com pindec s assoc iated w ith issuesof dicultie s has a careerindecisionand re he ercouess in g , e n it of decisionnsive asseis one of perm the ssm a s ts the central makingment of d ieren m ajor im po for clients reith dierenttypesof diculties rtance in ca w n sel its developa l., 1 9 9 6 ; tial intervent ions (Gati e t Santos, 2 0 0 1). The ing the em taxono and thenality-rela my EPCD could assist rs the of locat- proposed isticnal oranxiety, and personalidentcounselo inWe bproce ssby a s se unlikes si-m otio otonal diculties ted sources of clients dicultiesthat ssin g p ely that em i views, perso diusion. elieve ng it is could be itydysfuncional personaity resolved w ithout counseli interventios of focus ecison proble ore er- m eijs and D e-Boec, 2002). Locating ing ion th e m (G chronic and t l anteceden of th e d tns ms k the sources can help tailorc the eorts inform aion- elated Theem otional diculties red on retrie- valal approach to to addrethethe m . diculties, w hich isdistinct from the typi and counselors ss clients se should be cente career t proces of career in approprate C oun s lors canand in tere ts. For elo a mself-e ste em rsveloy form r n. e elp w ith x d sing aspirations n g eaatiorie sof experien cestohencoclients clients w p letyoun s e lo ement p c h av e to a rra nce build for their engagingin th e s e re er decision -m develop m a sei ing before skills urag a ident aking process. i, and con- ith the anxie aroused by the processand c e n e e dto m akea choice should D ealing w de focus of th e counselin g pro ces . Specically, counselorsm ay focus on ty th also be a such clients w ith m anag their ans assisti m ajor ng as condu cting in ter-view s aim ed at collecting carry ou t seem revi wting ing xiety as they informa ing e taskssuch and interest inven tion and ly rou in e persona lity lts. monitoring clients se lf-e stecom ponentof thAs th e utresults show e tory resu ty through m and anxie e career-counseling p ed, actively o the careerdecision-m aking process should be an integral
Appendix In elf-conceio n s am o n gth ) 1 1 EPCD scal s (N = 7 4 7 ,PVPesimistic Vie w s , A. rco e Ax Anxiety, SI Ste rre la t and Identit e pt y EPCD Scale 1. PV about the proc s es 2. PV about the w orld of work 3. PV about the individuals control 4. Axabout the process 5. Axabout uncertainty 6. Axabout the choice 7. Axabout the outcomes 8 . S I g e nral anxiety e 9 . SI se s te e m lf-e 1 0 . SI id entity status 1 1 . SI s n i can t others ig 1 2 3 4 5 6 7 8 9 10 11

.2 9 .3 6 .4 7 .5 0 .5 2 .5 4 .4 2 .2 1 .2 3 .4 2 .1 8 .2 8 .3 5 .3 3 .3 5 .1 7 .1 9 .2 2 .1 3 .3 8 .4 6 .4 0 .3 7 .2 2 .2 9 .3 8 .2 1 .7 0 .6 9 .4 8 .3 9 .3 2 .4 3 .2 7

.7 5 .5 5 .4 2 .4 1 .5 5 .3 0

.6 8 .3 5 .3 6 .5 2 .2 6

.2 4 .3 0 .4 7 .2 4

.5 3 .3 6 .5 1 .3 3 .4 0 .4 3